Rozeboom et al reported a predictive model for intensive care unit admission following surgery. This can help to identify a patient who may require more aggressive management following surgery. The authors are from the University of Colorado.
Patient selection: prior to surgery
Parameters:
(1) type of surgery
(2) relative work units (WRU)
(3) patient status
(4) ASA status
(5) emergency surgery
(6) functional status
Parameter
Finding
Points
type of surgery
gynecology
-1.171
neurosurgery
1.015
orthopedics
-1.661
otolaryngology
1.008
plastic
-0.083
thoracic
0.854
urology
-1.238
vascular
1.668
general
0
work relative unit
0.086 * (WRU)
status
outpatient
0
inpatient
2.98
ASA status
ASA1
0
ASA2
1.250
ASA3
2.096
ASA4 or ASA5
2.905
emergency surgery
no
0
yes
0.846
functional status
independent
0
partially dependent
1.03
totally dependent
1.426
value of X =
= SUM(points of parameters) - 8.3515
probability of postoperative ICU stay =
= 1 / (1 + EXP((-1) * X))
Performance:
• The area under the ROC curve is 0.93.
To read more or access our algorithms and calculators, please log in or register.
